A recent Posner opinion on chicken-gutting illustrates how judges change details they don’t like.

By Garrett Epps

“Facts are stubborn things,” John Adams once said, “and whatever may be our wishes, our inclinations, or the dictates of our passions, they cannot alter the state of facts and evidence.”  

Adams, however, was not an appellate court judge. In jurists’ hands, facts and evidence sometimes are surprisingly compliant. That’s true even though a basic legal principle declares that on appeal, the court may not rely on anything but the record submitted. Appellate judges are supposed to decide law, leaving the facts to the courts below.

The indefatigable Josh Blackman recently called attention to an opinion by Judge Richard Posner of the Seventh Circuit. The employees of a chicken processing plant sought overtime pay for removing their protective gear (“sterilized jacket, plastic apron, cut‐resistant gloves, plastic sleeves, earplugs, and a hairnet”) at lunchtimes and putting it back on afterwards. The parties had disagreed about how long the process took, so Posner apparently videoed his clerks, or some other luckless underlings, doing it.

“The videotape” (which is not available to the public) “reveals that the average time it takes to remove the clothing/equipment is 15 seconds and the average time to put it on is 95 seconds,” Posner wrote.  The workers’ complaint, he concluded, was what lawyers call “de minimis,” a mere trifle.

In a separate opinion, Judge Diane Wood pointed out two problems. First, she said, appellate judges are really not supposed to do that. “To the extent (even slight) that the court is relying on this experiment to resolve a disputed issue of fact, I believe that it has strayed beyond the boundaries” established by the federal rules that dictate how courts operate. 

Beyond that, she noted, it wasn’t a fair test. Unlike judicial employees, poultry workers spend their shifts wallowing in raw chicken guts. They are required by law to wash thoroughly before going among humans. (Posner blew this off by saying that they “would doubtless do it without being told to.”)

The problem isn’t new. Posner’s Seventh Circuit colleague, Frank Easterbrook, stirred a similar controversy a quarter-century ago when he brushed aside a defendant’s apparently airtight alibi. Witnesses, relying on a hospital clock, testified that the defendant would not have had time to commit the crime. Easterbrook just reached back and reset the clock: “Suppose the clock at the Hospital was a few minutes fast (digital watches were rare in 1967).”

Nor is it confined to the Seventh Circuit. Earlier in April, Simon Lazarus detailed how Judge Raymond Randolph of the D.C. Circuit spouted misinformation from the bench about the “unmitigated disaster” and “sky-high” costs of the Affordable Care Act launch—talk-radio points that were not in the record and, indeed, were clearly false.

We shouldn’t be entirely surprised that judges are sometimes a little confused about whether facts are really all that stubborn. Sometimes the confusion stems from procedure. Many cases come to the Court without a trial; the “facts” are just claims, and specific rules require the judges to assume one or the other version is correct—even if that version seems ridiculous or implausible. 

In other cases, the parties strategically allow the record to mislead. In Flagrant Conduct: The Story of Lawrence v. Texas, for example, author Dale Carpenter details the reshaping of “facts” in a challenge to Texas’s law against same-sex sodomy. The two men charged with sodomy barely knew each other, and probably actually weren’t having sex when arrested. As it moved up the appellate chain, though, the story became one of two men consummating their love within a loving relationship. 

No one else is allowed to talk back to these judges; why should mere reality be an exception? In some cases, their opinions slowly become less factual and more, to adapt a concept from Stephen Colbert, “facty.”

For example, some judges seem to regard Knight Rider or 24 as part of the record. Plumhoff v. Rickard, argued March 4, concerned a woman shot to death by police while she was a passenger in a fleeing car. Justice Antonin Scalia scoffed at the idea that police shouldn’t shoot at a fleeing driver. “You think it is clearly established law, clearly established law, that you cannot shoot to kill a driver whose  . . . car is moving?” he asked lawyer for the woman’s family.  “You watch the movies about bank robberies, you know, it happens all the time. Are these movies unrealistic?” 

And then there are times when the justices seem to want to overrule the laws of mathematics.

I’ve begun to worry that some court somewhere will hold that Pasteur’s germ theory is “just a theory” because Blackstone didn’t believe it.

Consider the oral argument in Hall v. Florida. This case, argued March 3, tests the Court’s holding in 2002 that states cannot put “mentally retarded” defendants to death. After that decision, the Florida Supreme Court created a hard-and-fast rule: Any defendant who tests above 70 on an I.Q. test is not “retarded,” may not offer any clinical evidence of intellectual disability, and may be executed.

That rule—followed by eight states—defies the clinical definition of “intellectual disability” (the preferred term), which has three parts: a low IQ score, difficulty adapting to daily life, and onset of symptoms before age 18. An IQ score by itself doesn’t determine anything. That’s in part because designers of IQ tests know their test scores are subject to a “standard error of measurement,” or SEM, which reflects how closely the text reflects the test-taker’s hypothetical “true score.”   

Justice Scalia wondered aloud whether the SEM had been calculated “for the purpose of determining who is so incapable of controlling his actions that he shouldn’t be subject to the death penalty?”

The correct answer to that question is “wha’?” There’s no “liberal” SEM or “conservative” SEM. It’s a number. Statisticians reach it through a series of elaborate and careful computations.

Admittedly, the specifics of how they do this—involving square roots and standard deviations—might not be immediately obvious someone with a J.D. degree; I had to spend time on the phone and the Internet to figure it out. But each justice has four clerks, any one of whom could consult the four (page 18, footnote 24) authoritative sources cited in the amicus brief filed by the American Association on Intellectual and Developmental Disabilities. That brief is part of the record, and having a clerk check it is a good deal more legitimate than having one doff and don chicken-gutting clothes. 

I’ve begun to worry that the morning paper will report that some court somewhere will hold that Pasteur’s germ theory is “just a theory” because Blackstone didn’t believe it, or that five justices have overruled 186,000 miles per second as the speed of light, arguing in an opinion that “sixty years after Einstein’s death, things have changed dramatically.”